The work described in the Research Plan will have two important consequences. First, it will result in more efficient, less expensive methods for the production of accurate BAC fingerprint maps and end sequences. Second, it will result in the generation of at least 2,800,000 fingerprinted clones, assembled into fingerprint maps for 14 large (e.g. mammalian) genomes. We will undertake our proposed software and "wet lab" technology development exercises in the context of a production fingerprint mapping effort, using methods we established to generate the maps over the three years of the proposal. Software development will include automated lane tracking of fingerprinting gels, improving our automated restriction fragment identification software (BandLeader), and modification of it so that it produces probability-based measures of restriction fragment quality. These "quality values", produced for each restriction fragment, will be exploited by new software we are developing to correctly and automatically order clones within contigs. Automatic selection of clone tiling sets will also be developed. Fingerprint maps will be made available for download on our website in FPC format and for viewing using our Internet Contig Explorer (ICE) software. We will provide lists of tiling set clones to the sequencing centers to support their sequencing goals. BAC-end sequencing technology development will emphasize experiments aimed at reduction in volume of expensive sequencing reagents, less expensive DNA purification and reduced use of expensive plasticware. We will place special emphasis on improving capillary array electrophoresis and low volume thermocycling methods as these may impact both costs of sequencing reagents and DNA purification. Finally, we will evaluate commercially available BAC DNA purification reagents for performance in both end sequencing and fingerprinting. If the performance and cost of the reagents are favorable, we will undertake experiments to automate one or more of the commercially available kits. Should this be successful, we will integrate the new protocols and processes into our production fingerprinting group. [unreadable] [unreadable]